Geo-Tagging Digital Images

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, are described for automatically and/or interactively editing image metadata. In one aspect, the methods include the actions of obtaining two from among three items of geo-metadata associated with an image. The three items of geo-metadata include an acquisition geo-location for the image, an acquisition geo-orientation for the image and a geo-location of a subject depicted in the image. The methods further include the actions of representing on a map the two obtained items of geo-metadata associated with the image, and determining candidates for the third item of geo-metadata based on the representations of the two obtained items of geo-metadata. In addition, the methods can also include the actions of determining the third item of geo-metadata associated with the image from among the generated candidates.

What is claimed is: 
     
         1 . A method performed by one or more processes executing on a computer system, the method comprising:
 obtaining, from geo-metadata associated with an image, an acquisition geo-location for the image and, an acquisition geo-orientation for the image;   representing on a map the acquisition geo-location and the acquisition geo-orientation;   identifying one or more objects based, at least in part, on the acquisition geo-location and the acquisition geo-orientation, wherein the one or more objects represent one or more candidates of a subject depicted in the image;   identifying, a primary candidate, the primary candidate positioned at the distance nearest the geo-metadata associated with the image;   identifying the primary candidate as the subject of the image; and   representing the subject on the map.   
     
     
         2 . The method of  claim 1 , further comprising:
 overlaying on the map a vector having an origin at the obtained acquisition geo-location and a direction corresponding to the obtained acquisition geo-orientation; and   identifying the one or more candidates positioned along the direction of the vector based on geo-information corresponding to the one or more candidates.   
     
     
         3 . The method of  claim 2 , wherein identifying the primary candidate as the subject of the image comprises:
 determining which of the one or more identified candidates is the subject based on a user selection.   
     
     
         4 . The method of  claim 2 , wherein identifying the primary candidate as the subject of the image comprises:
 selecting a candidate from among the one or more identified candidates that is located within a specified range of the acquisition geo-location.   
     
     
         5 . The method of  claim 2 , further comprising tagging the image in accordance with hierarchical geo-location information associated with the determined subject. 
     
     
         6 . The method of  claim 2 , further comprising generating metadata for a focus distance, wherein the focus distance is substantially equal to a length of the overlaid vector representing the acquisition geo-orientation. 
     
     
         7 . The method of  claim 2 , wherein identifying the one or more candidates positioned along the direction of the vector further comprises:
 receiving a user request to identify an object depicted in the image along the direction of the vector;   extending, in response to the received user request, a length of the vector; and   identifying the object upon the extended vector reaching a polygonal contour of the object.   
     
     
         8 . The method of  claim 7 , wherein reaching the polygonal contour occurs when the extended vector is separated from the polygonal contour by less than a predetermined distance. 
     
     
         9 . The method of  claim 2 , where the image comprises a frame included in a sequence of video frames, and the method further comprises associating the subject with the sequence of video frames. 
     
     
         10 . The method of  claim 9 , further comprising:
 obtaining an acquisition geo-location and an acquisition geo-orientation of an other frame of the sequence of video frames;   representing on the map the obtained acquisition geo-location and geo-orientation of the other frame, at least in part, by overlaying on the map an other vector having an origin at the obtained acquisition geo-location of the other frame and a direction corresponding to the obtained acquisition geo-orientation of the other frame;   identifying along the direction of the other vector one or more other candidates of a subject depicted in the image based, at least in part, on geo-information corresponding to the one or more other candidates;   identifying an other primary candidate, the other primary candidate positioned at the distance nearest the geo-metadata associated with the image;   identifying the other primary candidate as the subject of the other frame; and   associating the subject of the other frame with the sequence of video frames.   
     
     
         11 . A non-transitory computer readable medium encoded with a computer program, the program comprising instructions that when executed by a data processing apparatus cause the data processing apparatus to:
 obtain from geo-metadata associated with an image, an acquisition geo-location for the image and an acquisition geo-orientation for the image;   represent on a map the acquisition geo-location and the acquisition geo-orientation;   determine one or more objects based, at least in part, on the acquisition geo-location and the acquisition geo-orientation, wherein the one or more objects represent one or more candidates of a subject depicted in the image;   identify a primary candidate, the primary candidate positioned at the distance nearest the geo-metadata associated with the image;   identify the primary candidate as the subject of the image; and   represent the subject on the map.   
     
     
         12 . The non-transitory computer readable medium of  claim 11 , wherein the instructions to cause the data processing apparatus to represent on a map the acquisition geo-location and the acquisition geo-orientation comprise instructions to cause the data processing apparatus to:
 overlay on the map a vector having an origin at the obtained acquisition geo-location and a direction corresponding to the obtained acquisition geo-orientation, and   wherein the instructions to cause the data processing apparatus to determine one or more objects comprise instructions to cause the data processing apparatus to identify the one or more objects positioned along the direction of the vector based on geo-information corresponding to the one or more candidates.   
     
     
         13 . The non-transitory computer readable medium of  claim 12 , wherein the instructions to cause the data processing apparatus to identify the primary candidate as the subject of the image comprises instructions to:
 determine which of the one or more identified candidates is the subject based on a user selection.   
     
     
         14 . The non-transitory computer readable medium of  claim 12 , wherein the instructions to cause the data processing apparatus to identify the primary candidate as the subject of the image comprises instructions to:
 select a candidate from among the one or more identified candidates that is located within a specified range of the acquisition geo-location.   
     
     
         15 . The non-transitory computer readable medium of  claim 12 , further comprising instructions to cause the data processing apparatus to tag the image in accordance with hierarchical geo-location information associated with the determined subject. 
     
     
         16 . The non-transitory computer readable medium of  claim 12 , further comprising instructions to generate metadata for a focus distance, wherein the focus distance is substantially equal to a length of the overlaid vector representing the acquisition geo-orientation. 
     
     
         17 . The non-transitory computer readable medium of  claim 12 , wherein the instructions to cause the data processing apparatus to identify the one or more candidates positioned along the direction of the vector comprises instructions to:
 receive a user request to identify an object depicted in the image along the direction of the vector;   extend, in response to the received user request, a length of the vector; and   identify the object upon the extended vector reaching a polygonal contour of the object.   
     
     
         18 . The non-transitory computer readable medium of  claim 12 , where the image comprises a frame included in a sequence of video frames, and the method further comprises associating the subject with the sequence of video frames. 
     
     
         19 . The non-transitory computer readable medium of  claim 18 , further comprising instructions to cause the data processing apparatus to:
 obtain an acquisition geo-location and an acquisition geo-orientation of an other frame of the sequence of video frames;   represent on the map the obtained acquisition geo-location and geo-orientation of the other frame, at least in part, by overlaying on the map an other vector having an origin at the obtained acquisition geo-location of the other frame and a direction corresponding to the obtained acquisition geo-orientation of the other frame;   identify along the direction of the other vector one or more other candidates of a subject depicted in the image based, at least in part, on geo-information corresponding to the one or more other candidates;   identifying an other primary candidate, the other primary candidate positioned at the distance nearest the geo-metadata associated with the image;   identifying the other primary candidate as the subject of the other frame; and   associating the subject of the other frame with the sequence of video frames.   
     
     
         20 . A system comprising:
 at least one display device;   at least one processor; and   storage configured to store instructions that are operable, when executed by the processor, to cause the system to perform operations comprising:   obtaining, from geo-metadata associated with an image, an acquisition geo-location for the image and, an acquisition geo-orientation for the image;   representing on a map the acquisition geo-location and the acquisition geo-orientation;   identifying one or more objects based, at least in part, on the acquisition geo-location and the acquisition geo-orientation, wherein the one or more objects represent one or more candidates of a subject depicted in the image;   identifying, a primary candidate, the primary candidate positioned at the distance nearest the geo-metadata associated with the image;   identifying the primary candidate as the subject of the image; and   representing the subject on the map.